The Duke University Medical Center proposes to continue as the treatment and therapeutic evaluation center for children with human immunodeficiency virus Infection in North Carolina and adjacent referral areas. An increasing number of infected children are being recognized and effective therapeutic modalities for children are being developed and tested in Phase I, II & III protocols. The pharmacokinetics of zidovudine has been studied in children at Duke and the appropriate dosage will be utilized in a multi-institutional, open label trial to demonstrate efficacy in comparison to historical controls. Pharmacokinetics of ZDV and ultimately of other compounds will be determined in newborns. A placebo versus ZDV trial in infants born to seropositive women is planned as attempted intervention. Sequential evaluation of babies will provide data concerning prospective rate of acquisition of infection. Comparisons will be made of infected babies and uninfected infants who have been randomized to receive ZDV. Development, cardiomyopathy, neurological function, pulmonary function and development of HIV associated infections will be monitored. A specific research project is being designed to delineate the human B cell repertoire in neonate born to seropositive mothers. An established limiting dilution culture system of EBV infected B cells will be utilized to study the ontogeny and diversity of B cell repertoire. Several prototype antibody systems will be used to determine if the B cell repertoire of an infant with exposure to HIV develops normally. Recognition of cells secreting of HIV specific antibodies may provide a means of early neonatal diagnosis. Another means of early diagnosis will be an evaluation of specific HIV directed cytotoxicity. addition, careful observations will be made of the phenotypic appearance, and functional behavior of the immune system in treated and untreated newborns during their first year of life.